Approximating Replication Lag in Cross-Zone Replicated Block Storage Devices

1. A system for managing replication lag within a virtualized block storage device that is replicated across at least first and second zones of a cloud provider network, wherein the first and second zones are in communication via a communication network but are electrically and physically isolated from one another, the system comprising: a first computing system associated with the first zone, the first computing system including one or more computing devices implementing a first replicated data volume of the virtualized block storage device, wherein the first replicated data volume is configured to obtain writes to the virtualized block storage device, individual writes being of a defined byte size and associated with sequence numbers indicating an ordering of the writes; and a second computing system associated with a second zone that is electrically and physically isolated from the first computing system, the second computing system including one or more computing devices implementing a second replicated data volume of the virtualized block storage device, the second replicated data volume configured to replicate the writes from the first replicated data volume; and one or more computing devices configured to: obtain, from the one or more computing devices implementing the first replicated data volume, a first sequence number identifying a last write to be persisted to the first replicated data volume at a point in time; obtain, from the one or more computing devices implementing the second replicated data volume, a second sequence number identifying a last write to be persisted to the second replicated data volume at the point in time; determine a difference in first and second sequence numbers; calculate an expected maximum data loss size during failure of the first replicated data volume at least partly by multiplying the difference in first and second sequence numbers by the defined byte size; and transmit the expected maximum data loss size during failure of the first replicated data volume to a client device associated with the virtualized block storage device.

2. The system of claim 1 , wherein writes to the virtualized block storage device are represented as log entries within a write journal.

3. The system of claim 1 , wherein each of the first and second replicated data volumes is implemented by at least two computing devices in the respective first and second zones, the at least two computing devices including a primary device configured to obtain writes at the respective first or second replicated data volume and a secondary device configured to replicate the writes obtained at the primary device.

4. The system of claim 1 , wherein the one or more computing devices are further configured to determine that the expected maximum data loss size during failure of the first replicated data volume exceeds a threshold value, and wherein transmission of the expected maximum data loss size is responsive to determining that the expected maximum data loss size exceeds a threshold value.

5. A computer-implemented method for managing replication lag within a virtualized block storage device that is replicated across at least first and second zones of a cloud provider network, wherein the first and second zones are in communication via a communication network but are isolated from one another, the method comprising: obtaining, from one or more computing devices implementing a first replicated data volume of the virtualized block storage device, a first sequence number identifying a last write to be persisted to the first replicated data volume at a point in time, wherein the first replicated data volume is configured to obtain writes to the virtualized block storage device, individual writes being of a defined byte size and associated with sequence numbers indicating an ordering of the writes; obtaining, from the one or more computing devices implementing a second replicated data volume of the virtualized block storage device, the second replicated data volume being configured to replicate the writes from the first replicated data volume, a second sequence number identifying a last write to be persisted to the second replicated data volume at the point in time; calculating a replication lag metric reflecting replication lag between the first and second replication data volumes at least partly by multiplying difference in first and second sequence numbers by the defined byte size; and transmitting the replication lag metric to metric data store device associated with the virtualized block storage device.

6. The computer-implemented method of claim 5 , wherein the metric data store device is a client device.

7. The computer-implemented method of claim 5 , wherein the replication lag metric represents an expected maximum data loss size during failure of the first replicated data volume.

8. The computer-implemented method of claim 5 , wherein the writes to the virtualized block storage device are encrypted.

9. The computer-implemented method of claim 5 , wherein the a defined byte size is a product of a number of blocks written to within individual writes and a block size on the virtualized block storage device.

10. The computer-implemented method of claim 5 further comprising: obtaining, from one or more computing devices implementing the first replicated data volume, a first time reflecting when a write identified by a third sequence number was persisted to the first replicated data volume; obtaining, from the one or more computing devices implementing the second replicated data volume, a second time reflecting when a write identified by the third sequence number was persisted to the second replicated data volume; calculating a second replication lag metric reflecting a difference between the first and second times, the second replication lag metric representing a time required to replicate writes between the first and second replicated data volumes; and transmitting the second replication lag metric to the metric data store device.

11. The computer-implemented method of claim 5 further comprising: generating a statistical value associate with the replication lag metric and at least one previous calculation of the replication lag metric; and transmitting the statistical value to the metric data store device.

12. The computer-implemented method of claim 5 further comprising repeatedly recalculating the replication lag metric during a set of additional points in time and, for each recalculated value, transmitting the recalculated value to the metric data store device.

13. The computer-implemented method of claim 5 , wherein the method is implemented by a computing system comprising devices distributed between the first and second zones.

14. Non-transitory computer-readable media comprising computer-executable instructions implementable on a computing system associated with a virtualized block storage device that is replicated across at least first and second zones of a cloud provider network, the first and second zones in communication via a communication network but isolated from one another, wherein the instructions, when executed by the computing system, cause the computing system to: obtain, from one or more computing devices implementing a first replicated data volume of the virtualized block storage device, a first sequence number identifying a last write to be persisted to the first replicated data volume at a point in time, wherein the first replicated data volume is configured to obtain writes to the virtualized block storage device, individual writes being of a defined byte size and associated with sequence numbers indicating an ordering of the writes; obtain, from the one or more computing devices implementing a second replicated data volume of the virtualized block storage device, the second replicated data volume being configured to replicate the writes from the first replicated data volume, a second sequence number identifying a last write to be persisted to the second replicated data volume at the point in time; calculate a replication lag metric reflecting replication lag between the first and second replication data volumes at least partly based on a difference in first and second sequence numbers and the defined byte size; and transmit the replication lag metric to a metric output location associated with the virtualized block storage device.

15. The non-transitory computer-readable media of claim 14 , wherein the computing system comprises a set of computing devices distributed among the first and second zones.

16. The non-transitory computer-readable media of claim 14 , wherein the instructions further cause the computing system to: obtain, from one or more computing devices implementing the first replicated data volume, a first time reflecting when a write identified by a third sequence number was persisted to the first replicated data volume; obtain, from the one or more computing devices implementing the second replicated data volume, a second time reflecting when a write identified by the third sequence number was persisted to the second replicated data volume; calculate a second replication lag metric reflecting a difference between the first and second times, the second replication lag metric representing a time required to replicate writes between the first and second replicated data volumes; and transmit the second replication lag metric to the metric output location.

17. The non-transitory computer-readable media of claim 14 , wherein the instructions further cause the computing system to: generate a statistical value associate with the replication lag metric and at least one previous calculation of the replication lag metric; and transmit the statistical value to the metric data store device.

18. The non-transitory computer-readable media of claim 14 , wherein the instructions further cause the computing system to determine that the replication lag metric exceeds a threshold value, and wherein transmission of the replication lag metric is responsive to determining that the replication lag metric exceeds a threshold value.

19. The non-transitory computer-readable media of claim 14 , wherein the instructions further cause the computing system to repeatedly recalculate the replication lag metric during a set of additional points in time and transmit each recalculated value to the metric data store device.

20. The non-transitory computer-readable media of claim 14 , wherein recalculation of the replication lag metric occurs at a first periodicity and transmission of each recalculated value occurs at a second periodicity.